Arvind

Founder | AI Developer | Data Scientist | Ex- @Adobe, @Microsoft, @IBM | Tennis fan :)

[Aishwarya Srinivasan @aiwithaish]

/build-with-AI Buildathon is coming to San Francisco during SF Tech Week!

We’re so excited to host this at The Gen Academy in partnership with Mem0, SVAI Hub, and Snowflake on October 5th

This is an exclusive event with a small, select group at SVAI Hub. We’re keeping it close-knit so you can spend the day building alongside other AI builders, turning your idea into something real.

And yes, we have some very exciting prizes up for grabs!

Seats are VERY limited. If you want to be in the room, apply now before the spots are gone!

https://t.co/FHDFHPMjdh

[Deedy @deedydas]

I was wrong about Sarvam.

When I wrote about them a year ago, I felt like the direction to train small "indic" language models was wrong. But boy, have they turned it around. They have the best text-to-speech, speech-to text, and OCR models for Indic languages, and that's actually really valuable. The pricing is very reasonable. And the website is not only beautifully designed but dirt easy to use. They're filling a well needed gap in the ecosystem and doing things big labs will probably never focus on to the fullest extent (at least in the short term). I don't know anything about the business, but there's a lot to appreciate about what they've build technologically and I can't remember the last time I felt that way about software products coming out of India. Well done.

[Harrison Chase @hwchase17]

thinking of multi agent architectures on a casual saturday morning

any thoughts on the below?

[Harrison Chase @hwchase17]

Deep agents in JS!

Great for JavaScript/typescript developers looking to build deep research, coding, or other long running autonomous agents!

[Avi Chawla @_avichawla]

8 RAG architectures all AI Engineers should know:

[Aadit Sheth @aaditsh]

One of the truest paragraphs ever written

[Ryan Petersen @typesfast]

The difference between who you are today and who you’ll be in 5 years is almost entirely made up of the books you read between now and then.

[Harrison Chase @hwchase17]

OpenAI recently released a guide on building agents which contains some misguided takes

There's a lot of FUD, confusion, hype, and noise around agents

I wrote a blog on how to think about agent frameworks. Includes:

Background Info
- What is an agent?
- What is hard about building agents?
- What is LangGraph?

Flavors of agentic frameworks
- “Agents” vs “workflows”
- Declarative vs non-declarative
- Agent abstractions
- Multi agent

Common Questions
- What is the value of a framework?
- As the models get better, will everything become agents instead of workflows?
- What did OpenAI get wrong in their take?
- How do all the agent frameworks compare?

[Arvind @ArvNarayan]

@MatthewBerman I think the value is that its a RAG system wrapped in an interface that lets you do a lot more than just chat

[Arvind @ArvNarayan]

@Akshat_World People can have a difference of opinion on politics/policies and yet live peacefully.

[Arvind @ArvNarayan]

LLMs may have been introduced to us through interfaces like ChatGPT. But many of our existing systems are not designed to accommodate the flexible, conversational outputs that LLMs typically produce.

This is where structured output shines -

https://t.co/k9x4nSq9Kn

[Sam Altman @sama]

by very popular demand, structured outputs in the API:

https://t.co/AbfzTp4LMF

[Arvind @ArvNarayan]

@karpathy Isn’t that roughly what Youtube is about ?

[Arvind @ArvNarayan]

Check out our AI-powered Dashboard and Analyzer in action! This tool generates insights from selected data and performs advanced analytical tasks:

https://t.co/91DvhkYNos

Interested in bringing your AI/Data vision to reality? Feel free to reach out: https://t.co/8tjlmTO0Fi

[Arvind @ArvNarayan]

The AI models and platform providers will face a challenging few years as they strive to realize economic value. Many valuations will be reset along the way, and the advantage will belong to the industry giants with deep pockets.

[Arvind @ArvNarayan]

AI costs are currently way ahead of the revenue!

However, this does not mean that the AI wave is not real.

[Arvind @ArvNarayan]

[1] It enables sustainable experimentation, ensuring you stay competitive as AI value emerges.

[2] It allows you to stay flexible by adopting leading models/tools and upgrading to superior, cheaper options as they arise.

[Arvind @ArvNarayan]

@kunalb11 An LLM can definitely serve as a mentor or consultant, helping founders think with structured frameworks.

Agentic solutions like Autogen or CrewAI might be the way to go.

[Arvind @ArvNarayan]

@Akshat_World Just political posturing. Probability of realisation = 0